| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| llama-cpp-python is the Python bindings for llama.cpp. `llama-cpp-python` depends on class `Llama` in `llama.py` to load `.gguf` llama.cpp or Latency Machine Learning Models. The `__init__` constructor built in the `Llama` takes several parameters to configure the loading and running of the model. Other than `NUMA, LoRa settings`, `loading tokenizers,` and `hardware settings`, `__init__` also loads the `chat template` from targeted `.gguf` 's Metadata and furtherly parses it to `llama_chat_format.Jinja2ChatFormatter.to_chat_handler()` to construct the `self.chat_handler` for this model. Nevertheless, `Jinja2ChatFormatter` parse the `chat template` within the Metadate with sandbox-less `jinja2.Environment`, which is furthermore rendered in `__call__` to construct the `prompt` of interaction. This allows `jinja2` Server Side Template Injection which leads to remote code execution by a carefully constructed payload. |
| Unauthenticated file upload allows remote code execution.
This issue affects UvDesk Community: from 1.0.0 through 1.1.3.
|
| generator-jhipster-entity-audit is a JHipster module to enable entity audit and audit log page. Prior to 5.9.1, generator-jhipster-entity-audit allows unsafe reflection when having Javers selected as Entity Audit Framework. If an attacker manages to place some malicious classes into the classpath and also has access to these REST interface for calling the mentioned REST endpoints, using these lines of code can lead to unintended remote code execution. This vulnerability is fixed in 5.9.1. |
| The RESTful Content Syndication plugin for WordPress is vulnerable to arbitrary file uploads due to missing file type validation in the ingest_image() function in versions 1.1.0 to 1.5.0. This makes it possible for authenticated attackers, with Author-level access and above, to upload arbitrary files on the affected site's server which may make remote code execution possible. This requires the attacker have access to a defined third-party server as specified in the settings, so it is unlikely that this will be exploitable by contributor-level users, and more likely to be exploited by administrators who also have access to the plugin's settings. |
| A vulnerability in the application software of multiple Radiometer products may allow remote code execution and unauthorized device management when specific internal conditions are met. Exploitation requires that a remote connection is established with additional information obtained through other means. The issue is caused by a weakness in the analyzer’s application software. Other related CVE's are CVE-2025-14095 & CVE-2025-14096. Affected customers have been informed about this vulnerability. This CVE is being published to provide transparency.
Required Configuration for Exposure: Affected application software version is in use and remote support feature is enabled in the analyzer. Temporary work Around: If the network is not considered secure, please remove the analyzer from the network. Permanent solution:
Customers should ensure the following:
• The network is secure, and access follows best practices.
Local Radiometer representatives will contact all affected customers to discuss a permanent solution.
Exploit Status:
Researchers have provided working proof-of-concept (PoC). Radiometer is not aware of any publicly available exploits at the time of this publication. |
| Sitecore Experience Manager (XM) and Experience Platform (XP) 10.4 before KB1002844 allow remote code execution through insecure deserialization. |
| The Envolve Plugin plugin for WordPress is vulnerable to arbitrary file uploads due to missing file type validation in the 'zetra_languageUpload' and 'zetra_fontsUpload' functions in all versions up to, and including, 1.0. This makes it possible for unauthenticated attackers to upload arbitrary files on the affected site's server which may make remote code execution possible. |
| pac4j is a security framework for Java. `pac4j-core` prior to version 4.0.0 is affected by a Java deserialization vulnerability. The vulnerability affects systems that store externally controlled values in attributes of the `UserProfile` class from pac4j-core. It can be exploited by providing an attribute that contains a serialized Java object with a special prefix `{#sb64}` and Base64 encoding. This issue may lead to Remote Code Execution (RCE) in the worst case. Although a `RestrictedObjectInputStream` is in place, that puts some restriction on what classes can be deserialized, it still allows a broad range of java packages and potentially exploitable with different gadget chains. pac4j versions 4.0.0 and greater are not affected by this issue. Users are advised to upgrade. There are no known workarounds for this vulnerability. |
| An issue was discovered in Trusted Firmware-M through 2.1.0. User provided (and controlled) mailbox messages contain a pointer to a list of input arguments (in_vec) and output arguments (out_vec). These list pointers are never validated. Each argument list contains a buffer pointer and a buffer length field. After a PSA call, the length of the output arguments behind the unchecked pointer is updated in mailbox_direct_reply, regardless of the call result. This allows an attacker to write anywhere in the secure firmware, which can be used to take over the control flow, leading to remote code execution (RCE). |
| The Kognetiks Chatbot for WordPress plugin for WordPress is vulnerable to arbitrary file uploads due to missing file type validation in the chatbot_chatgpt_upload_file_to_assistant function in all versions up to, and including, 1.9.9. This makes it possible for unauthenticated attackers, with to upload arbitrary files on the affected site's server which may make remote code execution possible. |
| sveltekit-superforms makes SvelteKit forms a pleasure to use. sveltekit-superforms v2.27.3 and prior are susceptible to a prototype pollution vulnerability within the parseFormData function of formData.js. An attacker can inject string and array properties into Object.prototype, leading to denial of service, type confusion, and potential remote code execution in downstream applications that rely on polluted objects. This vulnerability is fixed in 2.27.4. |
| Markdownify is a Model Context Protocol server for converting almost anything to Markdown. Versions below 0.0.2 contain a command injection vulnerability, caused by the unsanitized use of input parameters within a call to child_process.exec, enabling an attacker to inject arbitrary system commands. Successful exploitation can lead to remote code execution under the server process's privileges. The server constructs and executes shell commands using unvalidated user input directly within command-line strings. This introduces the possibility of shell metacharacter injection (|, >, &&, etc.). This issue is fixed in version 0.0.2. |
| Remote Code Execution vulnerabilities are present in ASPECT if session administra-tor credentials become compromised.
This issue affects ASPECT-Enterprise: through 3.08.03; NEXUS Series: through 3.08.03; MATRIX Series: through 3.08.03. |
| Command injection vulnerabilities in the underlying CLI service could lead to unauthenticated remote code execution by sending specially crafted packets destined to the PAPI (Aruba's Access Point management protocol) UDP port (8211). Successful exploitation of these vulnerabilities results in the ability to execute arbitrary code as a privileged user on the underlying operating system. |
| Dolibarr ERP/CRM versions <= 3.1.1 and <= 3.2.0 contain a post-authenticated OS command injection vulnerability in its database backup feature. The export.php script fails to sanitize the sql_compat parameter, allowing authenticated users to inject arbitrary system commands, resulting in remote code execution on the server. |
| A specific flaw exists within the Bluetooth stack of the MIB3 unit. The issue results from the lack of proper validation of user-supplied data, which can result in an integer overflow when receiving fragmented HCI packets on a channel. An attacker can leverage this vulnerability to bypass the MTU check on a channel with enabled fragmentation. Consequently, this can lead to a buffer overflow in upper layer profiles, which can be used to obtain remote code execution.
The vulnerability was originally discovered in Skoda Superb III car with MIB3 infotainment unit OEM part number 3V0035820. The list of affected MIB3 OEM part numbers is provided in the referenced resources. |
| The Versa Director SD-WAN orchestration platform which makes use of Cisco NCS application service. Active and Standby Directors communicate over TCP ports 4566 and 4570 to exchange High Availability (HA) information using a shared password. Affected versions of Versa Director bound to these ports on all interfaces. An attacker that can access the Versa Director could access the NCS service on port 4566 and exploit it to perform unauthorized administrative actions and perform remote code execution. Customers are recommended to follow the hardening guide.
Versa Networks is not aware of any reported instance where this vulnerability was exploited. Proof of concept for this vulnerability has been disclosed by third party security researchers. |
| Llama Stack prior to version v0.2.20 accepted unverified parameters in the resolve_ast_by_type function which could potentially allow for remote code execution. |
| insightsoftware Hive JDBC through 2.6.13 has a remote code execution vulnerability. Attackers can inject malicious parameters into the JDBC URL, triggering JNDI injection during the process when the JDBC Driver uses this URL to connect to the database. This can further lead to remote code execution. |
|
Team ENVY, a Security Research TEAM has found a flaw that allows for a remote code execution on the camera. An attacker could inject malicious into http request packets to execute arbitrary code. The manufacturer has released patch firmware for the flaw, please refer to the manufacturer's report for details and workarounds.
|
